Inorganic Chemistry, Vol. 17, No. 8, 1978 2067
Trihalo(2-(2’-pyridyl)quinoline)gold(III) (20) A. D. Norman and J. A. Slater, unpublished observation. (21) H. P. Braendlin and E. T. McBee, “Friedel Crafts and Related Reactions”, Vol. 111, G. Olah, Ed., Interscience, New York, N.Y., 1964.
(22) W. P. Neumann, “The Organic Chemistry of Tin”, Interscience,New
York, N.Y., 1964.
(23) (24) (25) (26) (27)
J. C. Calabrese and L. F. Dahl, J. Am., Chem. SOC.,93,6042 (1971). J. Potenza and W. N. Lipscomb, Inorg. Chem., 5, 1471 (1966). J. Potenza and W. N. Lipscomb, Inorg. Chem., 5, 1474 (1966). J. Potenza and W. N. Lipscomb, Inorg. Chem., 5, 1483 (1966). A. Tippe and W. C. Hamilton, Inorg. Chem., 8, 464 (1969).
Contribution from the Chemistry Department, University of Virginia, Charlottesville, Virginia 22901
Synthesis, Structure, and Properties of Trichloro- and Tribromo(2- (2’-pyridyl)quinoline)gold( 111) C. J. O’CONNOR and E. SI”*
Received October 6, 1977 The title complexes Au(pq)C13and Au(pq)Br3 (pq = 2-(2’-pyridy1)quinoline) have been synthesized and their crystal and molecular structures determined from X-ray data obtained by counter methods. The compounds are isomorphous and crystallize in the space group P212121,with Z = 4 in unit cells of dimensions a = 7.711 (7), b = 10.04 ( l ) , and c = 19.34 (4) 8, (Au(pq)Cl,) and a = 8.050 (4), b = 10.13 (2), and c = 19.52 (1) 8, (Au(pq)Br,). The structures were solved by the heavy-atom method and refined by full-matrix least squares to R = 6.5% (Au(pq)C13, 1511 reflections) and R = 5.1% (Au(pq)Br3, 907 reflections). The compounds consist of discrete diamagnetic five-coordinated molecules. The distribution of the ligands about the metal is approximately square pyramidal with the axial ligand displaced from the vertical and at a relatively great distance from the metal, so that they may also be described as strongly distorted square planar. The direction of the distortion is such as to reduce the energy of the high-spin (triplet) state but not enough to produce a high-spin ground state.
Introduction
Gold compounds have been of long-standing biological interest for use as drugs and catalysts in biological systems.’ The observation of five-coordinated complexes is important in the understanding of gold(II1) chemistry. Many reactions of gold(II1) are proposed to proceed via a five-coordinated intermediate. The slow kinetics2 of ligand exchange and substitution with gold(II1) compounds may be explained by the relative instability of the five-coordinated intermediates compared with the square-planar compounds. Five-coordinated intermediates are also important in the action of gold(II1) on certain amino acids. Gold(II1) has been shown to stereospecifically drive the oxidation of @‘)-methionine to @)-methionine S-s~lfoxide,~ a reaction which presumably proceeds via a five-coordinateintermediate with the amino acid chelated to the gold(II1). The oxidative attack is then confined to an approach that is sterically limited by the other ligands (chlorine) on the gold. Ring-substituted dipyridyl, phenanthroline, and terpyridyl ligands induce a variety of unusual structures with transition metals, including pseudotetrahedral, five-coordinated monomeric, or dimeric compounds of manganese(II), nickel(II), and copper(II).e8 Stable five-coordinated gold complexes have been obtained with two such ligands,9J0 (I) 2,2’-biquinolyl and (11) 2,9-dimethylphenanthroline,due to extreme ligand-ligand
”
‘H
‘H
I biq
H‘
I1 dmP
nation of these structures leads us to postulate that five-coordinated complexes could be formed with the much less sterically hindered ligand (111) 2-(2’-pyridy1)quinoline. We report here on the synthesis and crystal structures of the five-coordinated gold(II1) complexes trichloro- and tribromo(2-(2’-pyridyl)quinoline)gold(III), Au(pq)X3, X = Cl, Br. Experimental Section Preparation of Compounds. Au(pq)CI3. Sodium tetrachloroaurate(II1) in absolute ethanol was added to a stoichiometric amount of pq in benzene. The orange-yellow crystals precipitated slowly and were filtered, washed with benzene and ethanol, and dried. Crystals suitable for X-ray study were obtained by slow evaporation of a saturated solution of the compound in nitrobenzene. Au(pq)Br3. The preparation of the compound is as above using NaAuBr4. The product is formed as orange-red crystals. Sodium tetrabromoaurate(II1) was prepared by repeatedly dissolving NaAuC4 in concentrated HBr and then heating to dryness. Z-(Z’-Pyridyl)quinoline. This ligand was prepared by the method of Smirnoff.” o-Aminobenzaldehyde, freshly prepared by steam distillation,I2 in ethanol is mixed with a stoichiometric amount of 2-acetylpyridine. After a stoichiometricamount of NaOH was added as 1 N NaOH, the mixture was refluxed for 1 h on a steam bath. The mixture was decolorized with charcoal, filtered, and diluted with hot water until turbid. The compound crystallized on cooling in ice, was filtered, and recrystallized twice from methanol. Magnetic Susceptibility. Measurements were made on a superconducting susceptometer and the complexes were found to be diamagnetic. Crystal Data for Au(pq)CI> AuC1~N2Cl4HI0: mol wt 510; space roup P212121;2 = 4; a = 7.711 (7),b = 10.04 ( l ) , c = 19.34 (4) v = 1498 A3;pcalcd= 2.13 g cm-3, Pobd = 2.2 g cm-3; r(Mo Ka) = 100 cm-l; crystal dimensions (mm from centroid) (100) 0.32, (TOO) 0.32, (010) 0.40, (oio) 0.40, (001) 0.038, (ooi) 0.038, (110) 0.30, (TIO) 0.30, (1iO) 0.295, (110) 0.295; maximum and minimum transmission coefficients 0.58 and 0.26. Crystal Data for Au(pq)Br,. AuBr3N2C1.,HIO,mol wt 643; space roup P21212 ; 2 = 4; a = 8.050 (4), b = 10.13 (2), c = 19.52 (1) v = 1592 X 3 ; pcalcd= 2.68 g cm-3, Pobd = 2.7 g cm-3; r(Mo Ka) = 175 cm-’, crystal dimensionsX2IO) 0.07, (210) 0.07, (011) 0.05, (011) 0.05, (01T) 0.05, (Oil) 0.05; maximum and minimum transmission coefficients 0.32 and 0.18. For each crystal, the Enraf-Nonius program SEARCH was used to obtain 15 accurately centered reflections which were then used in the program INDEX to obtain approximate cell dimensions and an ori-
1; 1;
H 111 P9
steric crowding caused by the ring substituents. An exami0020-1669/78/1317-2067$01.00/0
0 1978 American Chemical Societv
C. J. O'Connor and E. Sinn
2068 Inorganic Chemistry, Vol. 17, No. 8, 1978
Figure 1. Stereoview of Au(pq)Br3. entation matrix for data collection. Refined cell dimensions and their estimated standard deviations were obtained from least-squares refinement of 28 accurately centered reflections. The mosaicity of each crystal was examined by the w-scan technique and judged to be satisfactory, but the Au(pq)Br3 crystal was by far the better one. Collection and Reduction of Data. Diffraction data were collected at 292 K on an Enraf-Nonius four-circle CAD-4 diffractometer controlled by a PDP8/M computer, using Mo K a radiation from a highly oriented graphite crystal monochromator. The 8-28 scan technique was used to record the intensities for all nonequivalent reflections for which 0 < 28 < 50" for Au(pq)CI3 and 0 < 20 < 48' for Au(pq)Br3. Scan widths (SW) were calculated from the formula SW = A B tan 8 where A is estimated from the mosaicity of the crystal and B allows for the increase in width of peak due to Kal-Kcu2 splitting. The values of A and B were 0.60 and 0.30" for Au(pq)Br3 and 0.90 and 0.35' for Au(pq)CI,. The calculated scan angle is extended at each side by 25% for background determination (BG1 and BG2). The net count is then calculated as N C = TOT - 2(BG1 BG2) where TOT is the integrated peak intensity. Reflection data were considered insignificant if intensitia registered less than 10 counts above background on a rapid prescan, such reflections being rejected automatically by the computer. The intensities of four standard reflections, monitored for each crystal at 100 reflection intervals, showed no greater fluctuations during the data collection than those expected from Poisson statistics. The raw intensity data were corrected for Lorentz-polarization effects (including the polarization effect of the crystal monochromator) and then for absorption. The validity of the absorption correction was confirmed by rotating about the diffraction vector ("# scans") for several strong reflections; this showed that the transmission coefficients varied as expected with path length for each crystal. After averaging the intensities of equivalent reflectidlis, the data reduced to 1779 indc iendent intensities for Au(pq)CI3 and 1505 for Au(pq)Br3, of whi h 151 1 and 907 for the two complexes respectively had F: > 3a(F:), where u(F2) was estimated from counting ~tatistics.'~ These data were used in the final refinement of the structural parameters. Determination and Refinement of the Structures. The positions of the gold atom and three bromine atoms in Au(pq)Br3 were obtained from a three-dimensional Patterson function calculated from all intensity data. The intensity data were phased sufficiently well by these positional coordinates to permit location of the remaining nonhydrogen and some hydrogen atoms. Full-matrix least-squares refinement was based on F , and the function minimized was Cw(lFoI - lFc1)2. The weights w were then taken as [ ~ F , / U ( F ~ )where ] ~ , lFol and lFcl are the observed and calculated structure factor amplitudes. The atomic scattering factors for nonhydrogen atoms were taken from Cromer and Waber14 and those for hydrogen from Stewart et al." The effects of anomalous dispersion for all nonhydrogen atoms were included in F, using the values of Cromer and Ibers16 for Af' and Af". Agreement factors are defined as R = CllFol - IFcll/CIFoI and R, = (E:w(lF,! ~ F c ~ ) 2 / ~ w I F , ~ 2Refinement )1/2. on Au(pq)C13 was carried out using the parameters for Au(pq)Br3 as starting values. Four reflections were found to be suffering from secondary extinction effects and were eliminated from the data set of Au(pq)C13. Anisotropic temperature factors were introduced for all nonhydrogen atoms. Further Fourier difference functions permitted location of
+
+
the remaining hydrogen atoms, which were included in the refinement with fixed temperature factors (5.0 A*). The high-energy radiation used predetermines that the right and wrong choices for the absolute configuration will not greatly affect the refinement of the two complexes and therefore that the choice cannot safely be made on the intensities of a few reflections. For each complex, both configurations were refined to completion on all intensity data. The absolute configuration was chosen on the basis of slightly better agreement based on R, and the choice was confirmed by a detailed comparison of observed and calculated intensities of both weak and strong reflections. The models converged with R = 6.5, R, = 7.5% for Au(pq)CI3 and R = 5.1, R, = 5.4% for Au(pq)Br3. A structure factor calculation with all observed and unobserved reflections included (no refinement) give R = 7.3 and 6.7% for the two complexes, respectively; on this basis it was decided that careful measurement of reflections rejected automatically during data collection would not significantly improve the results. A final Fourier difference function was featureless. Tables of the observed structure factors are available." The principal programs used are as previously described.I8
Results and Discussion Stable, neutral complexes of pq are formed with gold(II1) halides. However, the precipitation of the compounds Au(pq)C13 and Au(pq)Br3 takes place over several hours, even though the compounds are insoluble in the reaction mixtures. This contrasts with the very rapid precipitation of other transition metal complexes with pq in similar reactions. The slow reaction is due to the relative instability of the fivecoordinated complex relative to the square-planar starting material, being forced to completion only by t h e removal of the insoluble product. This would relate directly to the slow process of ligand exchange and substitution reactions observed with gold(II1) compound^.^ Final positional and thermal parameters for Au(pq)Cl, and Au(pq)Br3 a r e given in Table I. Tables I1 and I11 contain the bond lengths and angles. The digits in parentheses in the tables are t h e estimated standard deviations in the least significant figures quoted and were derived from the inverse matrix in the course of least-squares refinement calculations. The light a t o m values for bond lengths, etc., have relatively high standard deviations d u e to the large scattering power accounted for by t h e heavy atoms. Because of t h e close structural similarity between the two complexes, only Au(pq)Br3 is shown in the diagrams. Figure 1 is a stereopair view of Au(pq)Br, a n d Figure 2 shows the molecular packing in the unit cell. The ligand environment of the gold atom is distorted square pyramidal in each case, with the gold atom raised above the base of the square pyramid by 0.04 A in Au(pq)C13 and 0.08 A in Au(pq)Br3. The main distortion from square-pyramidal geometry in each structure is t h e elongated A u - N ( l ) bond (2.66 (2) 8, in Au(pq)Br3 and 2.68 (1) A in Au(pq)C1,), and the tilt of the axis of the pyramid from the vertical, as defined
Inorganic Chemistry, Vol. 17, No. 8,I978 2069
Trihalo(2-(2’-pyridyl)quinoline)gold(III)
Table I. Positional and Thermal Parameters and Their Estimated Standard Deviations A. Au(pq)Cl, atom
X
Y
Z
Au Cl(1) Cl(2) CK3) atom
0.2995 (1) 0.587 (1) 0.014 (1) 0.314 (1)
0.4290 (1) 0.4316 (9) 0.4163 (12) 0.6408 (8)
0.66939 (6) 0.6463 (5) 0.6976 (6) 0.7110 (6)
N(1) N(2) C(2) C(3) C(4) C(5) C(6) C(7) C(8) C(9) C(10) C(2’) C(3‘)
0.233 (3) 0.292 (4) 0.305 (4) 0.367 (4) 0.362 (5) 0.298 (5) 0.226 (3) 0.291 (5) 0.211 (6) 0.152 (5) 0.156 (3) 0.297 (4) 0.303 (4)
atom Au Br(1) Br(2) Br(3) atom
Y
z
0.415 (3) 0.231 (3) 0.311 (3) 0.286 (3) 0.383 (4) 0.498 (4) 0.518 (3) 0.622 (4) 0.737 (4) 0.749 (5) 0.644 (3) 0.197 (3) 0.063 (3)
0.534 (1) 0.632 (1) 0.517 (1) 0.451 (2) 0.403 (2) 0.421 (2) 0.488 (1) 0.374 (2) 0.397 (2) 0.461 (2) 0.506 (1) 0.569 (1) 0.551 (2)
X
0.2999 (2) 0.5961 (4) 0.0074 (4) 0.3160 (6)
Y 0.4219 (1) 0.4190 (5) 0.4106 (6) 0.6376 (4)
X
Y
0.243 (3) 0.297 (4) 0.306 (4) 0.36 3 (4) 0.361 (5) 0.295 (4) 0.238 (4) 0.290 (5) 0.215 (6) 0.140 (6) 0.154 (4) 0.306 (4) 0.312 (5)
0.420 (3) 0.238 (3) 0.312 (3) 0.293 (4) 0.396 (4) 0.510 (3) 0.518 (4) 0.620 (4) 0.735 (5) 0.742 (5) 0.641 (4) 0.199 (3) 0.071 (4)
X
EllQ
0.67095 (7) 0.6484 (2) 0.7001 (2) 0.7184 (2) z
BZ2 1.26 (3) 2.4 (3) 6.0 (5) 1.4 (3) atom
E33
BIZ
3.73 (4) 7.6 (5) 6.9 (5) 7.8 (5)
0.04 (4) -0.3 (3) -0.4 (4) 0.4 (3) Y
X
B. Au(pqlBr3 Bl la B22 1.81 (4) 2.44 (4) 2.1 (1) 3.5 (2) 7.8 (3) 1.8 (1) 4.9 (2) 3.1 (2) B, A 2 atom
B33 2.74 (4) 5.5 (2) 5.7 (2) 5.8 (2) X
0.538 (1) 2.1 (5) 0.311 (5) 0.636 (1) 2.3 (5) 0.299 (5) 0.518 (2) 1.9 (6) 0.297 (4) 0.453 (2) 2.6 (8) 0.38 (3) 0.409 (2) 4.0(10) 0.34 (3) 0.427 (2) 2.5 (7) 0.30 (3) 0.493 (2) 2.9 (8) 0.21 (3) 0.383 (2) 4.3 (9) 0.09 (3) 0.406 (2) 5.6 (11) 0.15 (4) 0.471 (2) 6.1 (13) 0.34 (4) 0.513 (2) 3.0 (8) 0.3 1 (4) 0.569 (2) 2.3 (7) 0.29 (3) 0.548 (2) 0.30 (3) 3.3 (7) a The form of the anisotropic thermal parameter is exp[-1/4(B,Ih2u*2t B2zk2b*z t B3,11c*’ Table 11. Bond Lengths (A) for Au(pq)Cl, and Au(pq)Br, X=C1 X = B r X=C1 X = B r Au-X(l) Au-X(2) Au-X(3) Au-N(l) Au-N(2) N(l)-C(2) N(l)-C(6) N(2)-C(2‘) N(2)-C(6’) C(2)-C(3) C(2)-C(2’) C(3)-C(4)
2.261 (4) 2.271 (4) 2.277 (4) 2.68 (2) 2.11 (2) 1.22 (2) 1.36 (2) 1.26 (1) 1.42 (2) 1.40 (2) 1.52 (2) 1.34 (2)
2.422 (3) 2.423 (4) 2.385 (3) 2.64 (2) 1.99 (2) 1.27 (3) 1.32 (3) 1.36 (3) 1.40 (3) 1.37 (3) 1.52 (3) 1.35 (4)
C(4)-C(5) C(5)-C(6) C(5)-C(7) C(6)-C(10) C(7)-C(8) C(8)-C(9) C(9)-C(10) C(2’)-C(3’) C(3’)-C(4’) C(4’)-C(5’) C(5’)-C(6’)
1.30 (2) 1.43 (2) 1.54 (2) 1.42 (2) 1.38 (2) 1.33 (2) 1.37 (2) 1.39 (2) 1.34 (2) 1.34 (2) 1.38 (2)
1.32 (4) 1.38 (4) 1.41 (4) 1.47 (4) 1.38 (4) 1.42 (4) 1.31 (4) 1.37 (4) 1.44 (4) 1.31 (4) 1.32 (3)
by the angle N(l)-Au-N(2) (70.2 (9)’ in Au(pq)Br, and 68.0 (4)O in Au(pq)C13). The axial Au-N bond, which averages 2.67 8, in the two very similar complexes, is fairly weak (covalent Au-N = 1.40 0.74 = 2.04 A) but still much shorter than the van der Waals contact distance (nonbonded Au-N = 2.2 + 1.5 = 3.7 A).19 Axial distortion of the normal planar gold(II1) geometry toward square pyramidal lowers the energy of the lowest lying triplet state with respect to the singlet ground state. Such distortion observed in the fivecoordinated complexes Au(biq)Cl,, Au(dmp)C13, and Au(dmp)Br3 was insufficient to affect the magnetic properties; the compounds were observed to be diamagnetic.l0 The axial
+
-0.032 (4) 0.006 (5) 0.137 (3) 0.1960 0.3684 0.6097 0.8198 0.8357 0.6493 0.0287 -0.1176 - 0.0739 0.1678
0.304 (5) 0.298 (5) 0.296 (4) 0.4045 0.4041 0.3302 0.2043 0.0936 0.0969 0.3097 0.3111 0.2843 0.3417
3.1 (5) 3.5 (5) 2.6 (5) 2.8 (5) 4.6 (8) 4.2 (7) 2.1 (5) 4.7 (8) 5.3 (9) 5.0 (9) 2.0 (5) 2.1 (4) 3.5 (6)
Z
~~
2.65 (3) 2.9 (3) 2.4 (3) 5.8 (4) B, A’
BIZ 0.01 (7) -0.3 (2) -0.3 (2) -0.0 (2) Y
B23
B13
0.03 (5) 0.3 (3) 0.1 (3) 0.4 (5)
-0.47 (4) -0.9 (4) -1.7 (5) -2.0 (3) B, A’
z
0.599 (2) 0.666 (2) 0.686 (1) 0.4435 0.3558 0.3228 0.36 64 0.4748 0.5588 0.5009 0.5912 0.7048 0.7413
4.4 (7) 5.0 (8) 2.8 (5) 3.5 3.5 3.5 3.5 3.5 3.5 3.5 3.5 3.5 3.5 B23 -0.65 (8) -0.3 (2) -2.9 (3) -1.7 (2) B, A 2
B13
0.08 (6) 0.3 (1) 0.7 (2) 0.4 (2) Z
-0.030 (4) 0.599 (2) 0.004 (3) 0.663 (2) 0.130 (3) 0.680 (2) 0.20 (3) 0.43 (1) 0.40 (2) 0.38 (1) 0.59 (3) 0.34 (1) 0.77 (3) 0.39 (1) 0.79 (3) 0.49 (1) 0.56 (2) 0.65 (3) 0.49 (1) 0.08 (3) -0.12 (3) 0.59 (1) -0.06 (2) 0.70 (1) 0.16 (2) 0.71 (1) + 2B,,hka*b* t 2Bl,hla*c*
+
3.3 (8) 3.4 (7) 2.4 (6) 2.7 1.6 1.5 2.6 2.7 5.9 5.8 4.8 2.6 2.6 2B2,klb*c*)].
distortion is even weaker in the present complexes Au(pq)Br3 and Au(pq)C13, so that a singlet ground state would be predicted, and is confirmed by the observed diamagnetism of the compounds. In each of the known five-coordinated gold(II1) complexes (Table IV), the elongation of the axial bond and its angular displacement from the vertical (2) axis minimizes the interaction with the dzz orbital. Thus the stereochemistry adopted is such as to minimize the possibility of high-spin ground states. Steric constraints prevent more than one halogen sharing the plane of the very sterically hindered bidentate ligands biq (I) and dmp (111). Thus, the preferred planar geometry of gold(II1) can most nearly be achieved by bonding to one of the nitrogen atoms of I and 11. The approximate coplanarity of the three halogen atoms and one of the ligand nitrogens automatically brings the other ligand nitrogen to within bonding distance of the metal atom. The metal-ligand bond to this axial nitrogen atom will be weakened by the relative instability of five-coordinated gold(II1) and strong steric interaction of substituent protons adjacent to this nitrogen and the nearest coordinated halogen atom.1° In pq, the steric constraint is less severe, but still results in a five-coordinated complex. The great importance of steric interactions in distorting the various five-coordinated geometries is verified by the fact that the quinoline nitrogen atom forms an elongated axial bond and the pyridine ring is bonded in-plane. However,
2070 Inorganic Chemistry, Vol. 17, No. 8, 1978
C. J. O’Connor and E. Sinn
Figure 2. Molecular packing in Au(pq)Br3.
Table 111. Bond Angles (dog) for Au(pq)Cl, and Au(pq)Br, x = c1 X=Br X(lbAu-X(2) 176.4 (2) 175.4 (2) x(ij-Au-xi3j 90.7 iij 91.6 iij X( l)-AU-N( 1) 89.4 (4) 89.7 (3) X( l)-Au-N( 2) 86.3 (7) 88.2 (3) 90.3 (2) X( 2)-Au-X( 3) 90.8 (2) X(2)-Au-N( 1) 93.7 (4) 92.6 (3) X( 2)-Au-N( 2) 91.5 (7) 90.2 (3) X( 3)-Au-N( 1) 113.5 (6) 113.9 (3) X( 3)-Au-N( 2) 176.3 (6) 178.6 (4) N( 1)-Au-N( 2) 69.6 (8) 67.0 (4) Au-N( 1)-C(2) 103 (2) 102.2 (7) Au-N( 1)-C(6) 130 (2) 127.2 (8) C(2)-N ( 1)-C(6) 118 (2) 119.9 (7) Au-N( 2)-C( 2’) 127 (2) 125,9 (8) Au-N( 2)-C(6’) 122 ( 2 ) 111.8 (7) 111 (2) C( 2‘)-N( 2)-C(6‘) 122 (1) 123 (1) 122 (3) N( 1)-C( 2)-C( 3) 117 (2) N( 1)-C( 2)-C( 2’) 117 (1) 119 (1) 121 (3) C(3)-C(2)-C(2‘) Table IV. Comparative Bond Distances of Five-Coordinated Gold(II1) Complexes (A) Au(biq)-
Au-N (axial) Au-N (plane) Au-C1 (trans) (Au-Cl) (cis)
C1, 2.40 2.28 2.09 2.37
AUAu(dmp)- ( d m ~ ) C1, Br, 2.58 2.61 2.09 2.08 2.27 2.40 2.28 2.41
Au-
AU-
(W-
M-
C1, 2.68 2.11 2.28 2.27
Br, 2.64 1.99 2.39 2.42
the elongation of the axial bond is greater in the Au(pq)X3 complexes than in Au(biq)Cl, or the Au(dmp)X3 complexes. The most dramatic difference is between Au(pq)X3 (average 2.67 A) and Au(biq)C13 (2.40 A), in which the biq ligand is rotated such as to markedly shorten the axial Au-N bond. This rotation would minimize the interaction of the hydrogen atom adjacent to the in-plane nitrogen with the metal dZ2 orbital; the absence of a similar hydrogen atom in pq is the most significant difference between the two ligands and is a likely cause of the structural differences between the metal environments in the biq and pq complexes. The present results show that the unusual five-coordinated gold(II1) environment can be sterically forced by the choice of ligands and that this can be achieved with a ligand (pq) which has markedly less severe steric requirements than biq. However, the lesser steric constraint in the pq complex results in a weaker (longer) fifth metal-ligand bond than in the biq complex. The observation of stable five-coordinate gold(II1) complexes is very strong evidence in support of the proposed five-coordinate mechanisms of the chemistry of gold(II1) reactions.
x= c1 120 ( 1 ) 118 ( i j 121 (1) 126 (1) 114 (1) 118 (1) 122 (1) 120 (1) 121 (1) 122 (2) 121 (2) 123 (1) 121 (1) 116 (1) 116 (1) 121 ( 1 ) 118 (2) 123 (2) 115 (1)
-
X=Br
__
119 (3) 121 i 3 j 115 (3) 123 (3) 121 (3) 125 (3) 119 (3) 117 (3) I19 (3) 121 (4) 119 (3) 123 (3) 125 (2) 124 (3) 121 (2) 118 (3) 119 (3) 120 (4) 127 (3)
The only cases of five-coordinated gold(II1) occur when a fifth position is sterically forced, This indicates an instability of five-coordinated relative to four-coordinated geometry in the absence of other influences. However, the existence of stable five-coordinated gold(II1) complexes provides strong support for such a geometry in a reaction transition state. Thus ligand exchange and substitutions in aqueous and nonaqueous solutions tend to occur stepwise via five-coordinated intermediates but proceed very slowly because of the greater stability of the competing square-planar On the other hand, it seems unlikely that substitution and exchange reactions proceed via energetically unfavored and never observed three-coordinated intermediates. This is further illustrated by the observation that Me2AuOI-I prefers to form a hydroxy-bridged tetramer to achieve square-planar coordination rat her than show t hree-coordination ,2 Acknowledgment. Support received for instrumentation under NSF Grant GP-41679 is gratefully acknowledged. Registry No. Au(pq)Cl,, 66767-29-1; Au(pq)Br3, 66767-28-0; NaAuCl,, 15189-51-2; NaAuBr,, 52495-41-7. Supplementary Material Available: Listings of structure factor amplitudes (1 1 pages). Ordering information is given on any current masthead page.
References and Notes (1) P. J. Sadler, Struct. Bonding (Berlin), 29, 170 (1976). (2) L. Cattalini, A. Orio, and M. L. Tobe,J,Am. Chem.Soc., 89,3130 (1967). (3) E. Bodignon, L. Cattalini, G.Natile, and A. Scatturin, J . Chem. Soc., Chem. Commun., 878 (1943). (4) H. R. H. Patil, C. M~Harris, and E. Sinn, inorg. Chem., 8, 101 (1969). ( 5 ) E. Sinn, J . Chem. SOC.,Dalton Trans., 162 (7976). (6) R. J. Butcher and E. Sinn, J . Chem. SOC.,Dalton Trans., 1186 (1976).
Inorganic Chemistry, Vol. 17, No. 8, 1978 2071
Structure of AsC1,dmit (7) R. J. Butcher and E. Sinn, Inorg. Chem., 16, 2334 (1977). (8) R. J. Butcher, R. L. Shaw, and E. Sinn, unpublished work. (9) R. J. Charlton, C. M. Harris, H. R. H. Patil, and N. C. Stephenson, Inorg. Nucl. Chem. Lett., 2, 409 (1960). (10) W. T. Robinson and E. Sinn, J . Chem. SOC.,Dalton Trans., 726 (1975). (11) A. P. Smirnoff, Helv. Chim. Acta, 4, 802 (1921). (12) “Organic Syntheses”, Collect. Vol. 3, Wiley, New York, N.Y., revised, 1955, p 56. (13) P. W. R. Corfield, R. J. Doedens, and J. A. Ibers, Inorg. Chem., 6, 197 (1967). (14) D. T. Cromer and J. T. Waber, “International Tables for X-Ray Crystallography”,Vol. IV, Kynoch Press, Birmingham, England, 1974.
(15) R. F. Stewart, E. R. Davidson, and W. T. Simpson, J . Chem. Phys., 42, 3175 (1965). (16) D. T. Cromer and J. A. Ibers, ref 14. (17) Supplementary material. ( 18) D. P. Freyberg, G. M. Mockler, and E. Sinn, J. Chem. Soc., Dalton Tram., 441 (1916). 1 . -,
(1 9) L. Pauling, “The Nature of the Chemical Bond, 3rd ed,Cornell University Press, Ithaca, N.Y., 1960. (20) F. Coletta, R. Ettorre, and A. Sambarao, Inorg. Nucl. Chem. Lett., 8, 667 (1972). (21) G. E. Glass, J. H.Konnert, M. G. Miles, D. Britton, and R. S. Tobias, J . Am. Chem. SOC.,90, 1131 (1968).
Contribution from the Departments of Chemistry, Baldwin-Wallace College, Berea, Ohio 44017, and Georgetown University, Washington D.C. 20057, and the Office of Naval Research, Arlington, Virginia 22217
Crystal and Molecular Structure of TrichloroC1,3-dimethyl-2(3H)-imidazolethione]arsenic(111) DANIEL J. WILLIAMS,* CARL 0. QUICKSALL, and KENNETH J. WYNNE Received November 22, 1977
The crystal and molecular structure of AsC1,dmit (dmit = 1,3-dimethy1-2(3H)-imidazolethione)has been determined. The com ound, C3H8AsCI3N2S,crystallizes in space group P 2 , / c with four formula units in a cell of dimensions a = 8.787 (2) %,b = 10.207 (1) A, c = 12.723 (3) A, and @ = 108.41 (3)’. The calculated and observed densities are 1.898 and 1.89 (1) g ~ m -respectively. ~, The structure of AsC1,dmit consists of pseudo-trigonal-bipyramidalmolecules (sulfur equatorial) weakly joined into dimeric units of Ci symmetry via axial chlorine bridging. This is in contrast to the known structure of AsC13N(CH3), which is best described as a pseudo trigonal bipyramid with the amine bonding in the axial position.
The majority of structural work on adducts of group SA trihalides has been on complexes with Lewis bases that possess donor sites of relatively high electronegativity and small size such as nitrogen’,*and ~ x y g e n . From ~ , ~ the idealized structureS for AX3LE (A = central atom, X = halogen, L = two-electron donor, E = lone pair of electrons), a trigonal-bipyramidal array of substitutents would be predicted assuming a stereoactive lone pair (E). Structures I and I1 (Figure 1) show the energetically favorable isomeric possibilities for an AX3LE system. Both A s C ~ ~ N(Me M ~= ~ methyl) ~ and SbC13NH2C6HS2 exhibit structure I due to the relatively high electronegativity and small size of nitrogen.’ Since all structural studies for AX3L adducts to date have been concerned with bases of relatively high electronegativity, we were interested in whether a Lewis base with a larger donor site of relatively low electronegativity such as sulfur would give rise to an adduct possessing structure 11. The arsenic trichloride adduct of 1,3-dimethyl-2(3H)-imidazolethione (dmit) was therefore prepared, and the crystal and molecular structures were determined. H
H
W
S
dmit Experimental Section The preparation of AsC1,dmit has already been reported? Crystals for X-ray study were grown from a saturated methylene chloride solution at room temperature. A suitable crystal was mounted in a glass capillary, and preliminary precession photography indicated space group P 2 1 / cfrom systematic absences h01 for I = 2n + 1 and OkO for k = 2n + 1. The cell dimensions were obtained using Mo K a l radiation (0.709 300 A) at 25 ‘C by a least-squares refinement to fit the k28 values for 10 high-angle reflections centered on the Picker * To whom correspondence should be addressed at the Division of Natural Science and Mathematics, Kennesaw College, Marietta, Ga. 30061.
FACS-1 four-circle diffractometer. The cell arameters are a = 8.787 (1) A, b = 10.207 (1) A, c = 12.723 (3) (3 = 108.41 (2)’, and V = 1083 A3. For a formula weight of 309.48 (C5H8AsC13N2S)and by assuming Z = 4, the calculated density is 1.898 g cm-) which agrees well with the observed density of 1.89 g cm-3 obtained by flotation in a mixture of n-hexane and dibromomethane. Collection and Reduction of Data. The same crystal used for the unit cell determination was used for data collection. The approximate dimensions were 0.45 mm X 0.65 mm X 0.35 mm with the [I051 direction coincident with the goniometer rotation axis. Diffractometer data were obtained using Zr-filtered Mo K a radiation by the 8-28 scan technique at a takeoff angle of 1.5’. The peaks were scanned at a rate of 1.Oo/min from 1.Oo on the low-angle side of the Kal peak to 1.0’ on the high-angle side of the Kaz peak. A scintillation counter was employed to count the diffracted beam, and a Ni foil attenuator was used whenever the count rate exceeded 8000 counts. Stationary-crystal, stationary-counter background counts of 10-s duration were taken a t each end of the scan. A unique data set was collected up to 28 = 55’. The intensities of three reflections were monitored as standards every 100 reflections, and no loss in intensity was observed. Intensities were corrected for background, and standard deviations were assigned according to the equations Z = C- 0.5(tc/tb)(B1 B2) and u(Z) = [C + 0.25(tc/tb)2(Bl + Bz) + (pZ)2]1/2, where C is the integrated peak count obtained in time t , and B1 and B2 are the background counts obtained in time Tb, all corrected for scalar truncation. A value of 0.05 was used for p . The data were also corrected for Lorentz and polarization effects and subsequently for absorption. A linear absorption coefficient of 41.7 cm-’ was calculated to give transmission coefficients ranging from 0.14 to 0.32. Of the 2518 unique reflections collected, 2049 with Z 2 2 4 1 ) were used in the refinement. Solution and Refinement of Structure. The structure was initially solved by direct methods using a data set collected with Cu K a radiation and uncorrected for absorption. The details of this data collection and structural solution are reported ~ e p a r a t e l y . ~The J~ positions of the nonhydrogen atoms derived from the data set taken with Cu K a radiation were used as starting positions in refinement of the data set described above. One cycle of isotropic and three cycles of anisotropic least-squares refinement yielded R1 = 0.043 and R2 = 0.059, where R I = CIIFoI - lFcll/xlF0l and R2 = [xw(lFol IFC1)*/xwF2]I/z. In this refinement the function minimized was xw(lFol - lFc1)2where IF,I and IF,/are the observed and calculated structure amplitudes and where weights, w, were taken as 4F?/$(F,).
1,
+
0020-1669/78/1317-2071$01.00/00 1978 American Chemical Society
